Stefan Mileschin

For computer intensive workloads

Tyan has launched the latest GPU server platforms that support the Nvidia V100S Tensor Core and Nvidia T4 GPUs for a wide variety of compute-intensive workloads including AI training, inference, and supercomputing uses.

Tyan's Vice President Danny Hsu said that AI was infusing into data centres and more organisations plan to invest in AI infrastructure that supports the rapid business innovation.

“Tyan’s GPU server platforms with Nvidia V100S GPUs as the compute building block enables enterprise to power their AI infrastructure deployment and helps to solve the most computationally-intensive problems.”

There are two make flavours based around Intel Xeons or AMD EPYC 7002.

The Thunder HX FT83-B7119features high density local storage within a 4U 10GPU server platform. The system is based on dual-socket 2nd Gen Inte Xeon Scalable Processors, supporting up to 10 Nvidia V100S or 20 T4 GPUs with 12 hot-swap 3.5" drive bays. The system provides a spare PCIe x16 slot in addition to the 10 double-wide PCIe x16 ones, and it supports high speed networking such as 100 Gigabit EDR InfiniBand or Ethernet. The chassis features tool-less drive trays for added ease of service.

https://fudzilla.com/news/network/50...-based-servers